refactor exercises for lower time complexity - #1

refactor exercises for lower time complexity

• calculateSumAndProduct: merge two O(n) passes into one (still O(n),
  already optimal, just halves the constant factor)
• findCommonItems: O(n*m) -> O(n+m) using a Set for lookups
• hasPairWithSum: O(n^2) -> O(n) using a Set of seen values
• removeDuplicates: O(n^2) -> O(n) using a Set for duplicate checks

Applied to both the JavaScript and Python versions of each exercise,
with brief Time/Space/Optimal complexity notes per the exercise brief.

implement data structures, caching, and precomputing

• linked_list.py: doubly-linked list, O(1) push_head/pop_tail/remove
• lru_cache.py: O(1) get/set via hashmap + doubly-linked list
• skip_list.py: new file, O(log n) expected insert/contains
• fibonacci.py / making_change.py: manual memoization (no @cache)
• common_prefix.py: sort + adjacent-pair comparison instead of O(n^2)
• count_letters.py: precomputed character set instead of O(n^2)

All test suites pass, including the large-scale ones (1M strings,
10M-char string, ways_to_make_change(9176)).